WHEN I AM RUNNING THIS CODE I AM RECEIVING ERROR IS CASE OF CONVERSION FROM LAB TO RGB. HERE I HAVE WORKED WITH B COMPONENT.HOW CAN I COMBINE 3 CHANNEL AND GER RID OF FROM ERR

C = imread('Lena.png');

lab = rgb2lab(C);

L=lab(:,:,3);

A=lab(:,:,1);

B=lab(:,:,2);

%imshow(L);

output_size = max(size(L));

D = 315.25;

dsensor3 = 0.1985;

[F3, sensor_pos3, fan_rot_angles3] = fanbeam(L,D,...

'FanSensorSpacing',dsensor3);

%imshow(F3);

Ifan3 = ifanbeam(F3,D,'FanSensorSpacing',dsensor3,'OutputSize',output_size);

% imshow(Ifan3);

%img_block=blockproc(F3,[4 4],@perdm_fun,'BorderSize',[1 1],'TrimBorder',true);

mat=4*ones(1,90);

newRow = zeros(1,size(F3,2));

newF3 = [F3(1:359, :); newRow; F3(360:end, :)];

H=mat2cell(newF3,mat,mat);

D = 2;

S = 360 ;

X = arrayfun(@(x) S, 1:D,'un',0);

M = zeros(X{:});

p1=mat2cell(M,mat,mat);

p=mat2cell(M,mat,mat);

s=mat2cell(M,mat,mat);

for i=1:90

for j=1:90

celmat=zeros(4);

b=(H{i,j});

[p1{i,j},p{i,j},s{i,j}] = perdm_fun(b);

end

end

finalkey=keygen();

ncell=encypt(p1,s);

modF3=cell2mat(ncell);

IfanmodF3 = ifanbeam(modF3,D,'FanSensorSpacing',dsensor3,'OutputSize',output_size);

%aid=lab([A(:),B(:),IfanmodF3(:)]);

%aid2=lab2rgb(aid);

encyptedLena = lab2rgb(IfanmodF3);%,'ColorSpace','adobe-rgb-1998');

figure,imshow(encyptedLena);

ERROR:

Error using images.color.ColorConverter/convert (line 218)

Mismatch between input size and the expected number of input color components.

Error in images.color.ColorConverter/evaluate (line 364)

out = convert(self, in, varargin{:});

Error in images.color.internal.Callable/parenReference (line 15)

[varargout{1:nargout}] = self.evaluate(varargin{:});

Error in lab2rgb (line 76)

rgb = converter(lab);

Error in perdm_Main (line 45)

encyptedLena = lab2rgb(IfanmodF3);%,'ColorSpace','adobe-rgb-1998');

Answer 1

[Image Analyst]

I didn't go over the whole code (no time now) but I did notice something wrong immediately. L is channel 1, not 3. Corrected code:

rgbImage = imread('Lena.png');   
labImage = rgb2lab(rgbImage); 
[L, A, B] = imsplit(labImage);

[sabitri]

Thanks...solved this..my mistake....But now i am receiving this. My encryption portion code is given below @Image Analyst

for i=90:-1:1

for j=90:-1:1

smod1 = reshape(s{i,j}.',1,[]);

for l=1:16

S = dec2bin(typecast(double(smod1(l)), 'uint8'), 8) - '0';

F = dec2bin(typecast(double(finalkey(l)), 'uint8'), 8) - '0';

bi = xor(S, F);

ree=typecast(uint8(bin2dec(char(bi + '0'))), 'double');

if isnan(ree)

ree=0;

else

s1(l) =ree;

end

end

s11=reshape(s1,4,4);

ncell{i,j}=p1{i,j}+s11;

end

end

ERROR:

Error using reshape

Number of elements must not change. Use [] as one of the size inputs to automatically

calculate the appropriate size for that dimension.

Error in encypt (line 24)

s11=reshape(s1,4,4);

Error in perdm_Main (line 43)

ncell=encypt(p1,s);

[Walter Roberson]

            if isnan(ree)
               ree=0;
            else
               s1(l) =ree;
            end

in the nan case you assign to ree but you do not assign into s1.

[sabitri]

Sorry, i couldn't understand....can you explain please? @Walter Roberson

[Walter Roberson]

Suppose that this is the first time through the loop. You do not assign anything to s1 before

for i=90:-1:1

so the only way that anything gets assigned to s1 is in the else clause of the isnan test.

The isnan test is inside the for l=1:16 loop. Suppose isnan is false for the first 10 iterations of l, then with the else clause you assign to each of s1(1) to s1(10). Now suppose isnan is true when l is 11. You assign 0 to ree but you do not assign anything to s1(11) because the assignment is inside the else test. Now suppose isnan is false for l = 12, so you assign something to s1(12), which has the side effect of filling in 0 for the missing s1(11). Now keep going and assume for l=15 that isnan is false so you assign to s1(15), filling in any missing elements with 0. At this point s1 is a vector of length 15.

Now suppose that for 16 the value is nan, so you assign 0 to ree but you do not assign ree to s1(16). At this point s1 is still length 15.

Now you try to reshape the vector of length 15 to 4 by 4 and that fails.